The invention relates to a process for preparing triethylenetetramine (TETA) by hydrogenation of ethylenediaminediacetonitrile (EDDN) over a catalyst. If appropriate, EDDN can also be present as a constituent of an amino nitrile mixture which additionally comprises ethylenediaminemonoacetonitrile (EDMN).
It is generally known that aliphatic nitrites, which may be substituted by further functional groups, can be hydrogenated in the presence of catalysts to form the corresponding amines. As indicated below, such hydrogenation processes are also known for various amino nitrites for producing some amines. However, it has up to the present not been stated anywhere that TETA can also be prepared from the amino nitrite EDDN or, if appropriate, from an amino nitrite mixture comprising EEDN and EDMN by direct hydrogenation of the amino nitrile. The processes known hitherto for preparing TETA are, however, as indicated below, associated with disadvantages.
Numerous processes for hydrogenating the α-amino nitrites aminoacetonitrile (AAN) and iminodiacetonitrile (IDAN) or β-amino nitrites have been described in the prior art. Thus, it is known that the hydrogenation of β-amino nitrites generally proceeds without problems, while the hydrogenation of α-amino nitriles is associated with the occurrence of numerous disadvantages such as hydrogenolysis of the C—CN bond or the R2N—C bond. “Handbook of Heterogeneous Catalytic Hydrogenation for Organic Synthesis, pp. 213 to 215” indicates the problems of the hydrogenation of α-amino nitrites for the example of α-alkylamino nitrites or cyclic α-amino nitrites compared to β-amino nitrites. The known stability problems of α-amino nitrites are presumably the main reason why to the present day only the hydrogenation of the α-amino nitrites AAN or IDAN to EDA (ethylenediamine) or DETA (diethylenetriamine), respectively, has been described in detail. However, EDA or DETA are prepared industrially by means of the EDC or MEOA process described below. However, a corresponding hydrogenation is not known for higher α-amino nitrites.
DE-A 3 003 729 describes a process for hydrogenating aliphatic nitrites, alkyleneoxy nitrites and alkylene amino nitrites to primary amines of a cobalt or ruthenium catalyst in the presence of a solvent system. The solvent system used comprises water and ammonia together with an ether or polyether. The alkylene amino nitrites or alkylene oxy nitrites which can be used as starting materials are in each case defined by means of a complex general formula. As specific compounds or examples which can be hydrogenated to the corresponding diamine, mention is made of, inter alia, ethylenediaminedipropionitrile (EDDPN; also referred to as N,N′-bis(cyanoethyl)-ethylenediamine) or 3,3′-(ethylenedioxy)dipropionitrile. On the other hand, DE-A 3 003 729 does not make any reference to the use of individual compounds of EDA derivatives having cyanomethyl substituents, e.g. EDDN or EDMN. In addition, EDMN does not come within the general definition of alkylene amino nitrites according to this document.
EP-A 0 382 508 describes a process for the batchwise preparation of acyclic, aliphatic polyamines by hydrogenation of acyclic, aliphatic polynitriles in the liquid phase over Raney cobalt catalysts, preferably in the presence of anhydrous ammonia. Here, a polynitrile solution is fed into a reaction zone which comprises the Raney cobalt catalyst in an essentially oxygen-free atmosphere. During the entire time of the reaction, the polynitrile solution is fed in at a rate which is not greater than the maximum rate at which the polynitrile reacts with the hydrogen in the reaction zone. This process enables polyamines to be prepared from polynitriles such as iminodiacetonitrile (IDAN), nitrilotriacetonitrile, ethylenediaminetetraacetonitrile (EDTN) or further compounds which have 2 or more cyano groups and are not specified in more detail.
EP-A 212 986 relates to a further process in which the same aliphatic polynitriles as in EP-A 0 382 508 can be hydrogenated to the corresponding polyamines under a granular Raney cobalt catalyst in the presence of a liquid primary or secondary amine comprised in the feed stream. As amino components which have to be present, mention is made of, inter alia, ethylenediamine (EDA) and also numerous further primary or secondary amines.
EP-A 1 209 146 relates to a further process for the continuous hydrogenation of nitrites to primary amines, in which the respective nitrites are used in the liquid phase over a suspended, activated Raney catalyst based on an aluminum alloy and the reaction is carried out in the absence of ammonia and basic alkali metal or alkaline earth metal compounds. Nitriles which can be converted into the corresponding ethylene amines include, among many others, IDAN, EDTN, EDDPN or ethylenediaminemonopropionitrile (EDMPN).
EP-B 0 913 388 relates to a process for the catalytic hydrogenation of nitriles, which comprises contacting the nitrile with hydrogen in the presence of a cobalt sponge catalyst under conditions for carrying out the conversion of the nitrile groups into the primary amine. The cobalt sponge catalyst has been treated beforehand with a catalytic amount of lithium hydroxide and the process is carried out in the presence of water. Suitable nitrites are aliphatic nitrites having from 1 to 30 carbon atoms, including, inter alia, β-amino nitrites such as dimethylaminopropionitrile. A further process for preparing polyamines from the corresponding polynitriles is disclosed in DE-A 27 55 687. In this process, the hydrogenation is carried out over a pelletized hydrogenation catalyst in the presence of a stabilizer which inhibits decomposition of the catalyst. As polynitrile, it is possible to use, inter alia, ethylenediaminedipropionitrile (EDDPN). Suitable stabilizers include, inter alia, EDA.
US-A 2006/0041170 relates to a process for preparing TETA, in particular TETA salts, and their use as drugs. In this multistage process, EDDN is prepared first. EDDN is subsequently reacted with benzaldehyde to form a (cyclic) imidazolidine derivative. This cyclic compound, which has two cyano groups, is reduced, for example by reaction with hydrogen, to give the corresponding cyclic diamino compound. This diamino compound is in turn hydrolyzed in the presence of an acid to give the corresponding TETA salt. In an alternative embodiment, the cyclic diamine compound is likewise reacted with benzaldehyde to form the corresponding diimine compound which is subsequently once again hydrolyzed in the presence of an acid to give the corresponding TETA salt. As further process alternatives, this document describes the reaction of EDDN with Boc protective groups (tert-butoxycarbonyl groups). The resulting EDDN derivative protected by two Boc protective groups is subsequently hydrogenated to give the corresponding protected TETA derivative. The Boc protective groups are removed by acid hydrolysis to give the corresponding TETA salt. A disadvantage of the process described in US-A 2006/0041170 is, in particular, that it is a multistage hydrogenation process in which the starting material EDDN used firstly has to be chemically derivatized in order to carry out the hydrogenation. A further disadvantage is that TETA is initially obtained as salt but not in the free base form.
There is therefore no report anywhere in the prior art that EDDN or amino nitrile mixtures comprising EDDN or EDMN can be used for preparing TETA and, if appropriate, further ethylene amines by direct hydrogenation of the amino nitrile. However, other (industrial) processes for preparing TETA are known.
EP-A 222 934 relates to a process for preparing higher alkylenepolyamines by reaction of a vicinal dihaloalkane with an excess of ammonia in the aqueous phase with addition of a strong base, resulting in formation of an imine intermediate which is subsequently reacted with an alkylenepolyamine to form the higher alkylenepolyamine. A suitable vicinal dihaloalkane is, in particular, ethylene dichloride (EDC or 1,2-dichloroethane). Alkylenepolyamines used are, in particular, ethylenediamine or higher ethylene amines such as DETA and also TETA and tetraethylenepentamine (TEPA). These processes (EDC processes) give a mixture of various ethylene amines (linear ethylene amines such as EDA, DETA, TETA, TEPA or higher ethylene amines and also cyclic derivatives such as piperazine (Pip) or aminoethylpiperazine (AEPip)). Depending on the ethylene amine added to the starting materials EDC and NH3, the reaction mixture comprises a corresponding proportion of higher ethylene amines. If, for example, TEPA is to be specifically prepared, the ethylene amine TETA is added to the starting materials EDC and NH3. As a result, the product (ethylene amine mixture) comprises a relatively high proportion of TEPA, but also the abovementioned further linear and cyclic ethylene amines. Disadvantages of this process are, in particular, that the process proceeds with low selectivity (to the components of the ethylene amine mixture obtained) and that a specific ethylene amine (for example DETA) firstly has to be prepared and is subsequently introduced into the process to prepare the next higher ethylene amine (for example TETA) in a targeted manner or to increase the obtained yield. However, this process represents a corrosion problem because of the starting materials used (haloalkanes) and the hydrochloric acid formed and also an environmental problem because of the salts formed.
U.S. Pat. No. 3,462,493 relates to a process for preparing TETA, in which an at least five-fold molar excess of EDA is reacted with ethylene dichloride or ethylene dibromide. By-products obtained here are, in particular, Pip or piperazinoethylethylenediamine.
DE-T 689 11 508 describes an alternative process for preparing linearly extended polyalkylenepolyamines such as TEPA. In this process, a bifunctional aliphatic alcohol is reacted with an amine reactant in the presence of a tungsten-comprising catalyst. A suitable bifunctional aliphatic alcohol is, in particular, monoethanolamine (MEOA), an EDA or DETA, for example, can be used as amine reactants. This process in principle gives mixtures of linearly extended polyalkylenepolyamines (i.e. ethylene amine mixtures). These ethylene amine mixtures comprise DETA, TETA, TEPA, Pip, AEPip or piperazine derivatives of higher ethylene amines with the proportion of the respective components varying according to the amine reactants used. If DETA is used as amine reactant, an ethylene amine mixture having a high proportion of TETA and TEPA is obtained. Disadvantages of this process are that the process proceeds with low selectivity (to the components of the ethylene amine mixture obtained) and that an additional ethylene amine firstly has to be synthesized and then reacted with the bifunctional aliphatic alcohol (for example MEOA). This forms relatively large amounts of by-products such as aminoethylethanolamine (AEEA) or higher hydroxy-comprising ethylene amines which are of little economic interest. The relatively large amount of by-products obtained is due to the fact that MEOA or the higher ethanolamines (e.g. AEEA) can react with themselves instead of with the amine used. Owing to the (statistically) many reaction possibilities, the selectivity to the linear TETA is quite low because of the coproducts and cannot be controlled. The synthesis can be carried out only at a partial conversion.
A review of the preparation of ethylene amines is given by the SRI report “CEH Product Review Ethyleneamines”; SRI International, 2003; pp. 1-53, in which EDA or DETA, in particular, are prepared by the above-described processes (using the starting materials EDC or MEA). Here, higher ethylene amines such as TETA or TEPA are formed as by-products or are obtained in relatively high yield by further reaction of the starting materials with EDA or DETA.